top of page
Search

Understanding Well-Being in the Face of Large-Scale Crisis using Advanced Machine Learning Methods

In this study, we based our analyses on SPI's 4-wave comprehensive households data to reveal multileveled risk and protective factors to wellbeing during the COVID-19 pandemic.


Dr. Talia M. Schwartz-Tayri, Ph.D

Head and founder, the AI for SW lab

Spitzer Department of Social Work at Ben Gurion University of the Negev


Dr. Michael Fire, Ph.D

Software and Information Systems Engineering Department at Ben Gurion University of the Negev


Guy Navon

Software and Information Systems Engineering Department at Ben Gurion University of the Negev


Dr. Michal Grinstein-Weiss, Ph.D., Shanti K Khinduka Distinguished Professor

Washington University in St Louis, Saint Louis, MO



We based our analyses on SPI's 4-wave comprehensive households data to reveal multileveled risk and protective factors to wellbeing during the COVID-19 pandemic.

This project used an innovative machine learning approach to expose multileveled factors to explain resiliency during the COVID-19 pandemic and follows individuals' wellbeing changes throughout a year and a half.


Although recent resiliency research was groundbreaking, theoretical models have not incorporated the unique effects of multileveled factors, including the government policy, on individual and household resiliency.


To address this gap, we used SPI novel longitudinal 4-wave dataset (N=2,363) to reveal personal, familial, and policy-level features with the strongest predictive power on wellbeing. Data included comprehensive information on individual's and household sociodemographic characteristics, psychological wellbeing, emotional support, generalized anxiety, income loss due to COVID-19, nutrition insecurity, financial assets, government benefits, and more.


Full findings will be published soon.


Comments


bottom of page